OpenAI launched GPT-4o’s native image generation feature on March 25, 2025. This feature allows paying users (starting at $20 per month) to upload images and convert them into a Ghibli style or generate original images from text descriptions. This feature is also compatible with ChatGPT’s text-to-video model, Sora, which sparked a social media frenzy.
However, OpenAI’s servers are under tremendous pressure as user demand surges. CEO Sam Altman said on social media platforms, “Everyone’s love of images is awesome, but our GPUs are melting.” To deal with this challenge, OpenAI has decided to implement a temporary speed limit on ChatGPT’s image generation function to ensure stable system operation.
No official announcement has been made about when the free version will go live, so we may have to wait a while. However, stemming from the broad appeal of the Ghibli style and its expansion into the digital realm, users will be able to achieve similar results with the free tool even without a paid subscription.
Think ChatGPT Plus is too pricey? No problem! With these free tools, turn your selfies or pet photos into dreamy Ghibli-style portraits in just five minutes. Studio Ghibli’s soft tones and whimsical aesthetic have taken the world by storm, and now, AI is making it easy for everyone to experience it – without spending a dime. Follow the steps below to try it out!
Multiple Ways to Convert Photos to Ghibli-Style
Here are several common ways to convert photos to Ghibli Style (Ghibli), including technical principles and how to do it. In addition to OpenAI’s paid feature (ChatGPT Plus based on GPT-4o image generation), I will also provide cheaper or free alternatives.
Method 1: Use the Image-to-Image AI tool
Principle: After uploading a photo, AI re-renders the image in a Ghibli style based on a pre-trained model and cue words, retaining the original composition but adding features such as soft tones and dreamy backgrounds.
Tools and Steps:
- 01 Getimg.ai (free credits)
- Visit getimg.ai and upload a photo.
- Enter the prompt:
“Transform this photo into a Ghibli-style artwork, lush forest background, soft pastel colors, hand-painted feel.”Click “Generate,” and the result will be ready in 1-2 minutes.
The free quota is generated by about 50 monthly, which is enough for personal use. It’s free 50 times a month, enough to keep you entertained for a while.
- 02 PhotoLab (cheap)
- Download the PhotoLab app (iOS/Android). A free trial is available, but a subscription is required for premium filters (~$3/month).
- Choose “Anime” or “Cartoon” filter, adjust to Ghibli style (soft tones, natural background).
- Save the result.
- With this tool, the photo’s subject is still there; it’s just more of that dreamy Ghibli flavor.
This Method is suitable for personalized needs, preserving the main features of the photo.
Method 2: Using Text-to-Image AI Combined with Manual Adjustment
Principle: Use AI to generate a Ghibli-style background, then generate a close image by describing the people or objects in the photo, and finally synthesize it with editing software.
Tools & Steps:
- 01 Craiyon (totally free)
- Open craiyon.com.
- Enter description:
"A Ghibli-style girl with short black hair in a whimsical forest, soft sunlight, detailed textures."(Adjust the description according to the photo.) - After generating the image, download it.
- Cut and paste the face from the photo onto the generated image using free software (such as GIMP or Photopea) to adjust the blending.
This method is truly free but requires manual editing, which is suitable for users with basic editing skills.
Method 3: Using Filters and Post-Processing
Principle: Transform the photo into a cartoon style by using ready-made art filters and then manually adjusting the color tones and background to simulate a Ghibli style.
Tools and steps:
- 01 PicsArt (free + a small fee)
- Download the PicsArt app and upload photos.
- Use the “Cartoon” or “Sketch” filter to convert to cartoon style.
- Adjust the color tones (add soft greens and yellows) and add a natural background (e.g. forest, sky).
- Note: Free features are enough; premium material is about $5/month.
- 02 Canva (free)
- Upload photos to canva.com.
- Use
“Photo Effects” to convert to cartoon, then overlay Ghibli style background template(search ‘nature’ or “fantasy”). Fantasy”). - Adjust the transparency and color tone. Finished.
This method is simple and inexpensive, so it’s a good one to try quickly.
Method 4: Use an open-source AI model (Technology-Oriented)
Principle: Use a free, open-source image generation model (e.g., Stable Diffusion) to run and train locally in the Ghibli style.
Tools and steps:
- Stable Diffusion (free)
- Install Stable Diffusion on your computer (medium configuration required, graphics card recommended).
- Download the Ghibli style model (e.g., from the Hugging Face community).
- Upload a photo and enter the prompt: “Ghibli-style portrait, dreamy landscape, soft colors.”
- Generate and fine-tune.
This solution is completely free but requires a technical base, and the results are highly customizable.
Method 5: Use a cheaper or free AI generation tool
These tools can convert photos to Ghibli style using AI models:
- Artbreeder (partly free): You can generate artwork in Ghibli style via AI.
- Dream by Wombo (free & paid): Upload a photo and choose an “Anime” or “Illustration” style.
- Deep Dream Generator (partly free): You can upload a photo and apply an animated style filter.
- Runway ML (paid): Use Style Transfer or Gen-2 video generation.
- Fotor (partially free): Fotor offers AI style transfers, including anime filters.
Conclusion of Programs cheaper or free than OpenAI
- Getimg.ai: 50 times a month for free, premium version of 9 knives, foolproof operation, and more than OpenAI’s 20 knives to save a lot.
- Craiyon: Do not spend a penny. It is 30 seconds out of the picture, and the quality is almost enough.
- PicsArt: Free features enough to play, 5 dollars to unlock the full set, cheap and fast.
- Stable Diffusion: All are free and run their own models; the effect can be with OpenAI arm wrestling.
In general, the easiest way to use Craiyon, lazy people choose PicsArt, want to have a good effect on the Getimg.ai, technical geeks to get Stable Diffusion. Pick a photo, such as your dog or selfie, according to the steps to try, and in 5 minutes, you’ll have a whole Ghibli style!
Prompt Word Template For Creating Ghibli-Style Photos
The cue words are poorly written, and the diagrams that come out may be geeky; the photos are too complex, and the AI may not recognize them.
Here are some cue word templates to take directly to generate Ghibli Style (Ghibli Style) photos! These templates are suitable for various AI tools (such as Getimg.ai, Craiyon, and Stable Diffusion), whether you upload a photo to change the style or plain text to generate, they can be used. Just follow the set, and you’re guaranteed to have that soft, dreamy Ghibli flavor!
Template 1: Portrait (Photo to Ghibli)
BASIC VERSION: “Convert this photo into a Ghibli-style portrait with soft pastel colors, lush forest background, hand-painted textures, and soft sunlight.”
Variant 1: “Convert this image into a Ghibli-style artwork of a young girl with flowing hair surrounded by whimsical trees in dreamy blue and green tones.”
Variant 2: “Turn this photo into a Ghibli-style scene of a boy in a country village with warm golden tones and intricate hand-painted details.”
Template 2: Pet Photo Turned Ghibli
BASIC VERSION: “Change this pet photo into a Ghibli-style illustration, fluffy animal in a magical meadow, soft lighting, vibrant yet gentle colors.”
Variant 1: “Transform this dog picture into a Ghibli-style art, playful pup by a serene lake, pastel sunset glow, detailed nature backdrop.”
Variant 2: “Convert this cat image to a Ghibli-inspired scene, curious kitty on a mossy cliff, surrounded by floating lanterns, hand-drawn feel.”
Template 3: Panoramic scenes (with or without characters)
BASIC VERSION: “Render this photo as a Ghibli-style landscape, vast rolling hills, soft greens and yellows, whimsical clouds, hand-painted aesthetic.”
Variant 1: “Turn this image into a Ghibli-style panorama, person standing on a windy cliff, ocean waves below, pastel blues, detailed textures.”
Variant 2: “Transform this scene into a Ghibli-inspired artwork, ancient forest with glowing fireflies, gentle moonlight, and intricate foliage.”
Template 4: Fantasy elements (more dreamy)
BASIC VERSION: “Convert this photo to a Ghibli-style fantasy, a person with a mystical creature, surrounded by floating islands, soft purples and pinks.”
Variant 1: “Turn this image into a Ghibli-inspired magical scene, girl riding a giant bird, sky full of fluffy clouds, warm pastel tones.”
Variant 2: “Render this photo as a Ghibli-style wonderland, figure in a boat on a starry river, glowing water, hand-drawn details.”
Template 5: Plain text generation (no photo reference)
BASIC VERSION: “A Ghibli-style artwork of a young adventurer in a dense forest, soft sunlight filtering through trees, pastel colors, hand-painted look.”
Variant 1: “Ghibli-inspired portrait of an old fisherman by the sea, weathered face, stormy sky, gentle blues and grays, detailed background.”
Variant 2: “A Ghibli-style scene of a girl with a lantern in a snowy village, warm lights, soft whites and blues, and whimsical architecture.”
Tips for use
Details add up: Want to be more Kipling? Add words like “whimsical,” “soft lighting,” “hand-drawn,” or specific settings (forest, cliff, village). , village).
Color Adjustment: Use more words like “pastel,” “gentle,” and “soft” to emphasize the softness.
Tool adaptation: Getimg.ai starts directly with “Transform”, while Craiyon removes “photo” and describes it directly.
Notes On Turning Your Photos into Ghibli-Style
When using tools, make sure you’re mindful of your privacy: photo uploads are risky
You throw your photos to online AI tools (e.g. Getimg.ai, Craiyon), and the server may store your original image. Some companies will take that data to train models or accidentally leak it.
How to prevent this:
- Before you use it, take a look at the Privacy Policy to see what they do with uploaded images.
- Pick a reliable tool, such as Getimg.ai, that has a clear “no data for sale” statement.
- Blur sensitive photos (those with clear faces), or don’t upload them at all, and use text generation instead.
- Running Stable Diffusion locally is the safest way to keep your photos on your computer.
Tool limitations: do not exceed expectations
free tools (like Craiyon) to generate quality may be brown, low resolution; some have a limit (Getimg.ai 50 times a month), over the money out of pocket.
How to prevent:
- First, try a few waste pictures, feel the effect, and do not directly apply it to the precious photos.
- Tip: Try a few more times, adjust to satisfaction, and do not expect perfection.
- Super limit on the tool, such as Canva, PicsArt unlimited number of times.
Copyright: Who owns the generated image
Some tools (especially the free ones) may claim that they own the generated image or that you have to pay extra to use it.
How to prevent this:
- Check the “Terms of Use” (Terms of Use) to see what the copyright says.
- Free stuff like Craiyon is fine for personal use, but don’t sell it.
- Stable Diffusion’s own run of the map, the copyright belongs 100% to you.
Data security: The internet is not safe.
Uploading photos to go on the public internet can be intercepted by hackers, especially if you use unsecured Wi-Fi.
How to prevent this:
- Use your own Wi-Fi; don’t blindly pass them around in coffee shops.
- Websites must start with “https://”, if you don’t have this lock symbol, you should withdraw.
- Read the comments on mobile apps (like PicsArt) and don’t install anything from unknown sources.
As a final reminder, you can be bold enough to try these tools. Still, it is best to avoid uploading images that contain sensitive information, such as personally identifiable information (e.g., ID cards, passports), financial data, or other sensitive content, to prevent potential data breaches.
Images generated using AI may have copyright implications. For example, generating images that are similar in style to those of a particular artist may violate their copyright. Please ensure that you have the right to use the generated content to avoid infringing on the rights of others.
